#!/usr/bin/env python3
"""
自动生成 README 中的统计 badge 和表格

用法:
    python gen_stats.py              # 只打印统计
    python gen_stats.py --update-readme  # 更新 README.md 中的统计部分
"""

import json
import re
import argparse
from collections import Counter
from pathlib import Path

DATA_FILE = Path(__file__).parent / "data.jsonl"
README_FILE = Path(__file__).parent / "README.md"

LAYER_ORDER = ["混沌工程层", "基础设施层", "可观测层", "中间件层", "跨云与级联层", "应用层"]
LAYER_COLORS = {
    "混沌工程层": "red",
    "基础设施层": "blue",
    "可观测层": "purple",
    "中间件层": "orange",
    "跨云与级联层": "yellow",
    "应用层": "green",
}
DIFF_ORDER = ["easy", "medium", "hard"]
DIFF_COLORS = {"easy": "brightgreen", "medium": "orange", "hard": "red"}


def load_entries():
    entries = []
    with open(DATA_FILE) as f:
        for line in f:
            if line.strip():
                entries.append(json.loads(line))
    return entries


LAYERS = {"基础设施层", "应用层", "中间件层", "可观测层", "跨云与级联层", "混沌工程层"}
SUB_CATEGORIES = {
    "容器与工作负载", "节点与系统资源", "存储卷与持久化", "网络与DNS",
    "配置与密钥", "发布与回滚", "服务发现与入口", "探针与健康检查", "应用运行时",
    "缓存与Redis", "数据库", "消息队列", "CDN与网关",
    "监控与指标", "告警与路由", "日志与采集", "链路追踪与RUM",
    "多云一致性", "级联故障", "灾备与切换",
    "故障注入设计", "多故障与干扰", "题面与证据",
}
DIFFICULTIES = {"easy", "medium", "hard"}
SCOPES = {"single-cloud", "multi-cloud"}


def parse_tags(entry):
    """Parse tags from list format: [layer, sub_category, faults..., difficulty, scope]."""
    tags = entry["tags"]
    if isinstance(tags, dict):
        return tags
    # List format
    layer = [t for t in tags if t in LAYERS]
    sub_cat = [t for t in tags if t in SUB_CATEGORIES]
    diff = [t for t in tags if t in DIFFICULTIES]
    scope = [t for t in tags if t in SCOPES]
    faults = [t for t in tags if t not in LAYERS and t not in SUB_CATEGORIES and t not in DIFFICULTIES and t not in SCOPES]
    return {
        "layer": layer[0] if layer else "",
        "sub_category": sub_cat[0] if sub_cat else "",
        "faults": faults,
        "difficulty": diff[0] if diff else "",
        "scope": scope[0] if scope else "",
    }


def compute_stats(entries):
    parsed = [parse_tags(e) for e in entries]
    layer_counts = Counter(p["layer"] for p in parsed)
    sub_counts = Counter(p["sub_category"] for p in parsed)
    fault_counts = Counter()
    for p in parsed:
        for fault in p.get("faults", []):
            fault_counts[fault] += 1
    diff_counts = Counter(p["difficulty"] for p in parsed)
    return layer_counts, sub_counts, fault_counts, diff_counts


def build_stats_section(entries):
    layer_counts, sub_counts, fault_counts, diff_counts = compute_stats(entries)
    total = len(entries)

    lines = []
    lines.append(f"## Dataset Statistics\n")
    lines.append(f"**Total: {total} challenges**\n")

    # Layer badges
    lines.append(f"### By Layer\n")
    badge_line = " ".join(
        f"![{layer}](https://img.shields.io/badge/{layer}-{layer_counts.get(layer, 0)}-{LAYER_COLORS[layer]})"
        for layer in LAYER_ORDER
        if layer_counts.get(layer, 0) > 0
    )
    lines.append(badge_line + "\n")

    # Sub-category table
    lines.append(f"\n### By Sub-category\n")
    lines.append("| Sub-category | Count |")
    lines.append("|---|---|")
    for sub, count in sub_counts.most_common():
        lines.append(f"| {sub} | {count} |")

    # Difficulty badges
    lines.append(f"\n### By Difficulty\n")
    badge_line = " ".join(
        f"![{diff}](https://img.shields.io/badge/{diff}-{diff_counts.get(diff, 0)}-{DIFF_COLORS[diff]})"
        for diff in DIFF_ORDER
        if diff_counts.get(diff, 0) > 0
    )
    lines.append(badge_line + "\n")

    # Top faults
    lines.append(f"\n### Top Faults\n")
    lines.append("| Fault | Count |")
    lines.append("|---|---|")
    for fault, count in fault_counts.most_common(10):
        lines.append(f"| {fault} | {count} |")

    return "\n".join(lines)


def update_readme():
    readme = README_FILE.read_text()
    stats_section = build_stats_section(load_entries())

    # Replace between markers
    start_marker = "<!-- STATS_START -->"
    end_marker = "<!-- STATS_END -->"

    if start_marker in readme and end_marker in readme:
        pattern = re.compile(
            re.escape(start_marker) + r".*?" + re.escape(end_marker),
            re.DOTALL,
        )
        new_content = f"{start_marker}\n{stats_section}\n{end_marker}"
        readme = pattern.sub(new_content, readme)
    else:
        # Insert before Usage section
        usage_idx = readme.find("## Usage")
        if usage_idx > 0:
            readme = readme[:usage_idx] + f"{start_marker}\n{stats_section}\n{end_marker}\n\n" + readme[usage_idx:]
        else:
            readme += f"\n{start_marker}\n{stats_section}\n{end_marker}\n"

    README_FILE.write_text(readme)
    print(f"README.md updated with stats for {len(load_entries())} challenges")


def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("--update-readme", action="store_true", help="Update README.md")
    args = parser.parse_args()

    if args.update_readme:
        update_readme()
    else:
        entries = load_entries()
        layer_counts, sub_counts, fault_counts, diff_counts = compute_stats(entries)
        print(f"Total: {len(entries)} challenges\n")
        print("Layer distribution:")
        for l in LAYER_ORDER:
            print(f"  {l}: {layer_counts.get(l, 0)}")
        print("\nSub-category distribution:")
        for s, c in sub_counts.most_common():
            print(f"  {s}: {c}")


if __name__ == "__main__":
    main()